printer profiles - sending photos to a print lab

Hello, Marianne here with a question. I am trying to print my photos for the first time. I am planning to send them to an outside printing lab. I have calibrated my monitor so that part is taken care of. However, I am not sure how to prepare my photos for printing i.e. what to do with printer profile. I researched the issue online, but didn't find much on the subject. So here is what I understand - We get the printer (and paper) profile from the printing lab and install it on our computer. The next step is to soft proof in Photoshop using the printer profile. So this much I understand no problem. However, from what I read online it also implies that AFTER editing (with the help of soft proofing) we then ACTUALLY CONVERT our working file (which is in Adobe RGB in my case) to the printer profile using Edit/Convert to profile in Photoshop. We go to Edit/Convert to profile and in the "destination" dropdown menu we pick our printer profile. We check the box black point compensation (if needed) and pick the rendering intent (usually perceptual or relative colorimetric). Then we save the file as TIFF and we are ready to send it to the printing lab. Are there steps in the process that I am missing? Is there anything else involved? The video that I got the above information from was talking about the printing lab using a PRINTING PRESS to print, but what about if the printing lab uses inkjet printer just like the ones used to print at home, or what if the lab is using some other printing technique other that a PRINTING PRESS will the process of preparing the photo for printing on my end then be different than what I described above? I hope you can help me out with this. Thank you very much,Marianne

Hi, sounds like you have things about right if you want to be very finicky about the printing, AND IF your print lab supports this setup (a pro type lab will). On the other hand, a certain number of mainstream labs say (or at least did in the past) that they only take sRGB images, and that every image is treated as if it is sRGB; your prints would be screwed up in this case.

 

Something you may be missing is a color profile on your monitor? You said it is "calibrated," which is a different thing than "profiled." To profile a monitor, you need a piece of hardware that sits on the front of the screen, measuring color while a piece of special software displays various test colors. Then the software generates a monitor profile. If you haven't done this, there's not much point to being so finicky on the back end.

 

If I were you, I'd just adjust my AdobeRGB files, take a quick look at the softproof just in case anything "blew up," then save as a jpeg and send it. (Making sure, first, that the print house will accept non-sRGB images; if not, then "convert to" sRGB before sending.)

 

A lot of the photonet wizards will squawk about this, but it'll be pretty unlikely they have more experience in this sort of thing than I do. (I've made hundreds and hundreds of printer and camera profiles, some of which were used to make well over a million 8x10" portrait prints per week, and have done the troubleshooting on plenty of lab color problems.)

 

If you are gonna do real critical color work, you'll want to do test prints from your lab, evaluate under the proper type of lighting, then make fine corrections based on the print results. Given that these are your first prints, I can't imagine that you're gonna be that critical. But you might find that the prints are overall darker or lighter than you want; a lot of this depends on how bright (or dark) both your monitor and the print viewing conditions are. Bottom line is, get a couple test prints made before sinking a lot of money into the printing.

"The video that I got the above information from... "

 

What video was that?

If it didn't come from the lab you actually intend to use then forget it, and forget the print profile!

The printing lab will take care of that themselves.

 

All you need to do is ensure your colour-space is one that's expected by the printing lab/house. I.e. sRGB or AdobeRGB. It's unlikely they'll accommodate anything else, and even AdobeRGB may be too 'exotic' for them to handle.

 

If in doubt, contact the lab to see what colour-space they can handle. If you get the impression they don't know what on earth you're talking about - find a different lab or stick with sRGB. A lab aiming at the amateur market will only expect sRGB.

 

It's a moot point whether you'll see any difference in finished prints between sRGB and AdobeRGB anyway. The only difference is in deep shadow gamma and the extent of green it can show. If your pictures don't show highly saturated greens or cyans, then you'll likely see no difference between the two spaces.

 

BTW, the lab will almost certainly not be using a printing press, unless your pictures are to be used as long print-run book illustrations.

 

Converting your files to a printing press profile will almost certainly screw things up badly.

 

Just let the lab handle the printer profile. That's what their job is.

Edited March 22, 2018 by rodeo_joe|1

Hi, sounds like you have things about right if you want to be very finicky about the printing, AND IF your print lab supports this setup (a pro type lab will). On the other hand, a certain number of mainstream labs say (or at least did in the past) that they only take sRGB images, and that every image is treated as if it is sRGB; your prints would be screwed up in this case.

 

Something you may be missing is a color profile on your monitor? You said it is "calibrated," which is a different thing than "profiled." To profile a monitor, you need a piece of hardware that sits on the front of the screen, measuring color while a piece of special software displays various test colors. Then the software generates a monitor profile. If you haven't done this, there's not much point to being so finicky on the back end.

 

If I were you, I'd just adjust my AdobeRGB files, take a quick look at the softproof just in case anything "blew up," then save as a jpeg and send it. (Making sure, first, that the print house will accept non-sRGB images; if not, then "convert to" sRGB before sending.)

 

A lot of the photonet wizards will squawk about this, but it'll be pretty unlikely they have more experience in this sort of thing than I do. (I've made hundreds and hundreds of printer and camera profiles, some of which were used to make well over a million 8x10" portrait prints per week, and have done the troubleshooting on plenty of lab color problems.)

 

If you are gonna do real critical color work, you'll want to do test prints from your lab, evaluate under the proper type of lighting, then make fine corrections based on the print results. Given that these are your first prints, I can't imagine that you're gonna be that critical. But you might find that the prints are overall darker or lighter than you want; a lot of this depends on how bright (or dark) both your monitor and the print viewing conditions are. Bottom line is, get a couple test prints made before sinking a lot of money into the printing.

Hi,

thank you so much for your reply. It's nice to get a reply from a professional who made many printer profiles. My monitor is calibrated AND profiled. I used the hardware on top of the screen just as you described. Can I ask you - you said that you made a lot of printer profiles, but if the labs don't require the photographer to use and convert to printer profiles and are asking to submit sRGB then who did you make those printer profiles for? Thanks again,

Marianne

"The video that I got the above information from... "

 

What video was that?

If it didn't come from the lab you actually intend to use then forget it, and forget the print profile!

The printing lab will take care of that themselves.

 

All you need to do is ensure your colour-space is one that's expected by the printing lab/house. I.e. sRGB or AdobeRGB. It's unlikely they'll accommodate anything else, and even AdobeRGB may be too 'exotic' for them to handle.

 

If in doubt, contact the lab to see what colour-space they can handle. If you get the impression they don't know what on earth you're talking about - find a different lab or stick with sRGB. A lab aiming at the amateur market will only expect sRGB.

 

It's a moot point whether you'll see any difference in finished prints between sRGB and AdobeRGB anyway. The only difference is in deep shadow gamma and the extent of green it can show. If your pictures don't show highly saturated greens or cyans, then you'll likely see no difference between the two spaces.

 

BTW, the lab will almost certainly not be using a printing press, unless your pictures are to be used as long print-run book illustrations.

 

Converting your files to a printing press profile will almost certainly screw things up badly.

 

Just let the lab handle the printer profile. That's what their job is.

Thank you very much for your reply. It seems that you and the poster above are giving me the same message - forget the printer profiles. I am learning:)

Marianne

...but if the labs don't require the photographer to use and convert to printer profiles and are asking to submit sRGB then who did you make those printer profiles for?

 

Hi, I spent a lot of years with a large studio chain outfit in the U.S., where I did a lot of QC, R&D, and other technical work in the processing lab(s). For a few years I did much of the evaluations on potential new "small" printers, where small means usable in our studios. Part of this included making "good" color profiles; these machines would often be prototypes or early production models where the firmware had not been fully shaken out On a handful of printer models that made it into our actual studios I would sometimes act as an internal "product manager," making sure that they performed well in studio use. (You might be surprised at how thoroughly a tech support "help desk" can screw things up.)

 

I've made a lot of profiles just to study printer variability over time and across media; the data within the profiles can be compared to see exactly what colors are being affected.

 

I've also made a lot of input (camera) profiles in order to fine tune camera responses to what we wanted, as well as a number of experimental profiles to deal with certain production problems. Or sometimes, for certain purposes, you want printer profiles slanted towards certain viewing conditions; a number of times I've measured the spectral makeup of the viewing lamps, then used that data in the profiling software to tailor the prints to that exact lighting.

It's unlikely they'll accommodate anything else [than sRGB or AdobeRGB] and even AdobeRGB may be too 'exotic' for them to handle.

 

The sort of software used by large print labs can typically be configured to handle the color management any way they want. But if the lab doesn't strictly limit input profiles to sRGB then it's likely that they'll handle the full boat, including image files already converted to printer profiles.

 

But you almost have to run your own test to be sure.

 

As a note, I can think of one reason why a person might want to do the printer conversion themselves - it would allow them to select the rendering intent (relative colorimetric or perceptual). Otherwise the lab default, which might be either, would be used. But I don't think it would make much difference to most people.

It seems to me that an important consideration is the repeatability of the printing process.

 

If it is repeatable, you can get trial prints, profile them, then adjust the image file appropriately.

 

If a lab has more than one printer, then they might calibrate differently, and so, as you see it, not be repeatable.

 

My guess is that inkjet printers are pretty repeatable until you change ink or print heads.

 

(There will be a random noise component to all systems, but that can be averaged out.)

 

Optical (silver halide) printers should be repeatable if the optical system is appropriately calibrated, and the processing system stays close enough to standard.

"(There will be a random noise component to all systems, but that can be averaged out.)"

 

- Huh!? How do you get random noise in a bitstream of zeroes and ones?

 

It's the printing lab's job and responsibility to calibrate and profile its own printers.

 

All of the variables mentioned above should be taken care of by the print lab. Otherwise they're not doing their job properly.

 

Printer profiling need only be of concern if you're printing 'in-house' or on your own personal printer.

 

Conversion from one profile to another is lossy - you can only remove information from an image file, not restore it. Therefore conversion between printer profiles should only be done when absolutely necessary. It's far better to leave the file in a display space like sRGB or AdobeRGB until the printer output is known.

Edited March 26, 2018 by rodeo_joe|1

As I read through these posts I am reminded of one thing that I sometimes find confusing, the use of the words "prints" and "photographs" as synonyms. I myself would often say, " I made these prints last night in my darkroom." I guess it sounded classier than saying photographs. So, is the OP asking about prints made on an offset printing press or photographs made in a photo lab?

 

I would also like to know what her intent is. Does she need several dozen of each print or photograph for inclusion in a book or for mass sale or does she just want a few to hang on her walls? What lab is she using? MPIX?

"So, is the OP asking about prints made on an offset printing press or photographs made in a photo lab?"

 

- Good question. But I doubt offset litho printing is involved.

 

I'm guessing she's referring to having a photo-book or something similar printed. But it could just as easily be lightjet or poster-sized inkjet printing.

 

Whatever the printing medium, it should still be the lab's job to prepare the image files for whatever machinery they use. After all, you wouldn't expect to have to tell a car mechanic what size of plug or air or oil filter to fit in your vehicle.

Hi, I spent a lot of years with a large studio chain outfit in the U.S., where I did a lot of QC, R&D, and other technical work in the processing lab(s). For a few years I did much of the evaluations on potential new "small" printers, where small means usable in our studios. Part of this included making "good" color profiles; these machines would often be prototypes or early production models where the firmware had not been fully shaken out On a handful of printer models that made it into our actual studios I would sometimes act as an internal "product manager," making sure that they performed well in studio use. (You might be surprised at how thoroughly a tech support "help desk" can screw things up.)

 

I've made a lot of profiles just to study printer variability over time and across media; the data within the profiles can be compared to see exactly what colors are being affected.

 

I've also made a lot of input (camera) profiles in order to fine tune camera responses to what we wanted, as well as a number of experimental profiles to deal with certain production problems. Or sometimes, for certain purposes, you want printer profiles slanted towards certain viewing conditions; a number of times I've measured the spectral makeup of the viewing lamps, then used that data in the profiling software to tailor the prints to that exact lighting.

Hi, thank you for replying. You seem to have a lot of expertise and experience in the field, so I appreciate getting advice from you, it really helped so thank you:)

It seems to me that an important consideration is the repeatability of the printing process.

 

If it is repeatable, you can get trial prints, profile them, then adjust the image file appropriately.

 

If a lab has more than one printer, then they might calibrate differently, and so, as you see it, not be repeatable.

 

My guess is that inkjet printers are pretty repeatable until you change ink or print heads.

 

(There will be a random noise component to all systems, but that can be averaged out.)

 

Optical (silver halide) printers should be repeatable if the optical system is appropriately calibrated, and the processing system stays close enough to standard.

Thank you, that totally makes sense. I looked into several labs and some of them use silver halide, others digital printing press. I am getting my test prints from both. Will see what they look like.

 

(I wrote)

"(There will be a random noise component to all systems, but that can be averaged out.)"

 

- Huh!? How do you get random noise in a bitstream of zeroes and ones?

 

It doesn't stay as ones and zeros, or it doesn't look very nice.

 

If, for example, the digital values go through a DAC, and modulate the intensity of a laser, there will be random noise in the amplifiers and modulators.

 

For inkjet, the exact size of a droplet could vary for a variety of reasons that are random noise.

 

But there should be a mean value averaged over a large area.

 

But a new inkjet head, or new laser or laser driving electronics, would have a different calibration, that may or may not be corrected.